JTTEE5 20:486–496

DOI: 10.1007/s11666-010-9546-4
1059-9630/$19.00  ASM International
Peer Reviewed

Mechanical Property Mapping of Cold

Sprayed Ti Splats and Coatings
Dina Goldbaum, Richard R. Chromik, Stephen Yue, Eric Irissou, and Jean-Gabriel Legoux

(Submitted April 15, 2010; in revised form August 1, 2010)

Profile nanoindentation and nanoindentation mapping were used to investigate the mechanical prop-
erties of commercially pure cold spray Ti splats and coatings deposited at increasing deposition veloci-
ties. Three regions in the cold spray Ti splats have been indentified: the impact region, the jetting region,
and the upper splat region. The mechanical properties measured in these regions were tied to the cold
spray deposition process with help of optical and scanning electron microscopes. The jetting region was
observed to contribute to a metallurgical bonding of cold spray splats to cold spray splats and was
measured to have low hardness in comparison to the splat impact site and similar to the hardness in the
upper splat region. No increase in the profile coatings hardness with increase in the particle in-flight
velocity and coating thickness was found. A correlation between the mechanical properties and the
presently known deposition temperature, stress and dislocation density models was made.

Keywords cold spray, hardness mapping, nanoindentation,

titanium
1. Introduction
Cold spray is an emerging coating deposition process
(Ref 1) that is carried out at relatively low-temperature
condition (Ref 2, 3). The deposition is achieved by powder
acceleration to supersonic velocities. Upon the particle
impact, the material undergoes plastic deformation which
initially results in a conformal material adhesion (Ref 3).
At high enough deposition velocities, the particles
undergo adiabatic shear instability where a localized
fluctuation in the pressure and strain rate leads to a
localized rise in the material temperature followed by the
material jetting (Ref 4-8).
Much of the current understanding of the cold spray
process has been garnered by experimentation and mod-
eling of the deposition of Cu (Ref 5, 7-12). Significant
interest is placed on the temperature rise near the impact
site and especially in the jetting region as this can have a
profound effect on the final microstructure of cold spray
coatings. For Cu, a temperature rise of up to 1200 K was
estimated for the particle impacting at the deposition
velocity of 600 m/s (Ref 4). While the volume of research
on Ti is less, similar studies revealed details of the
interface phenomenon for this material (Ref 13-22).
Dina Goldbaum, Richard R. Chromik, and Stephen Yue,
Department of Mining and Materials Engineering, McGill
University, 3610 University Street, Montreal, QC H3A 2B2,
Canada; and Eric Irissou and Jean-Gabriel Legoux, National
Research Council Canada, Industrial Materials Institute, 75 de
Mortagne Blvd, Boucherville, QC J4B 6Y4, Canada. Contact
e-mail: richard.chromik@mcgill.ca.
Zahiri et al. (Ref 14) as well as Vlcek et al. (Ref 19)
reported a metallurgical material bonding that was
observed in the jetting region in cold sprayed Ti. Also,
Zahiri et al. (Ref 14) believed that high-temperature
fluctuations induced material softening and grain recrys-
tallization with preferential grain orientation in Ti cold
spray splats. Using transmission electron microscopy, Kim,
et al. showed both regions of high dislocation density at
the impact site of the splat with the substrate (Ref 23) and
regions of recrystallization (Ref 24). Thus, a complex and
varied microstructure is observed in cold sprayed Ti that
may also result in a variation in local mechanical properties.
In this study, the hardness mapping of cold sprayed Ti
splats and coatings was used to investigate the presence or
lack of thermal softening, which could be associated with
material melting in the jetting region. The work carried
out is based on a nanoindentation mapping of individual
Ti cold sprayed particles deposited at various deposition
velocity conditions. The variation in mechanical proper-
ties within individual cold spray splats and also within
coatings was determined. Nanoindentation technique was
used to provide a better understanding of the deposition
mechanisms of cold spray Ti splats deposited at various
deposition velocity conditions. The effect of the deposi-
tion velocity conditions on the mechanical properties of
the coatings was investigated.
2. Experimental Procedure
2.1 Sample Preparation
Cold Spray deposition was carried out with a Kinetic
4000 cold spray gun (CGT GmbH, Ampfing, Germany).
Nitrogen gas was used as the propelling gas and was
pressurized from 2 to 4 MPa and preheated to tempera-
tures ranging from 300 to 800 C. Feedstock powders were
introduced into the cold spray gun and passed through a

486—Volume 20(3) March 2011 Journal of Thermal Spray Technology


Fig. 1 Micrograph (SEM) of the CP Ti powder used in this
study
deLeval nozzle (MOC24, CGT, GmbH, Ampfing,
Germany), reaching in-flight particle velocities between
580 and 825 m/s. The particle velocity was measured in
free jet at various process conditions with a DPV2000
(Tecnar Automation, St-Bruno, QC, Canada) in cold
particle mode using a laser diode to illuminate the in-flight
particles. The deposition efficiency was measured by tak-
ing the ratio of the weight of the deposited coating over
the weight of the cold sprayed powder.
The feedstock powder was commercially pure titanium
produced by plasma atomization (Raymore Industries,
Inc., Boisbriand, QC, Canada). The Ti particles had a
spherical morphology (see Fig. 1) and an average particle
diameter of 29 lm. The powder size distribution and
powder microstructure were reported in a previous
publication (Ref 22).
The cold spray gun was mounted onto an automated
robotic arm, making possible a wide range of gun motions.
Two types of samples were produced by changing the
gunÕs traverse speed and powder feeding rate. For depo-
sition of coatings, the powder feeding rate was 20 g/min
and the gun traverse speed was kept at 0.33 m/s. Sixteen
coating layers were deposited on one line, on top of the
7.62 9 7.62 9 0.32 cm thick mild steel substrate. Single
splats were deposited on mild steel by changing the gun
traverse speed to 1.0 m/s and by decreasing the powder
feeding rate to 2.0-5.0 g/min. A third type of sample,
multi-pass splats, was produced by increasing the number
of gun passes over the substrate surface.
2.2 Microstructural Characterization
Cold spray splats were examined both in plan view and
in cross-section by scanning electron microscopy (SEM)
and light optical microscopy (LOM). For cross-section
examination, specimens were mounted in a cold-mount
epoxy. The mounted specimens were prepared with
metallographic techniques recommended for electron
backscatter diffraction (EBSD), which are designed to
minimize residual stresses induced by polishing damage.
Journal of Thermal Spray Technology
Peer Reviewed
Fig. 2 Micrograph of the cross-section of a Ti splat obtained by
LOM. The flattening ratio was measured from images such as
these to examine the extent of deformation of Ti splats as a
function of deposition velocity
The final polishing step was 0.05 lm colloidal silica. No
chemical etching was carried out to minimize surface
roughness. The cold spray coatings were prepared in the
same manner, first being cut across the coating cross-
section (perpendicular to the gun traverse direction). SEM
imaging was conducted in a variable pressure scanning
electron microscope (Hitachi S-3000 N VP-SEM, Japan).
Image analysis of SEM and LOM micrographs of cold
spray splats and coatings was carried out with commer-
cially available image analysis software (Clemex Vision
Professional 5.0, Clemex Technology Inc., Longueuil, QC,
Canada). LOM micrographs were used to measure the
splat flattening ratio, FR, by taking a ratio of splat width to
splat height (Ref 7, 25) as shown in Fig. 2. For each
deposition conditions, the flattening ratio of between five
and twenty-eight splats was measured. The coating
porosity was measured from SEM micrographs obtained
with the instrument in backscattered imaging mode at
2009 magnification. Over three micrographs were ana-
lyzed and the average of the measurements with a stan-
dard deviation was taken. While porosity measurements
quoted here are from SEM micrographs, measurements
conducted with LOM micrographs yielded similar results.
2.3 Nanoindentation Testing
The nanoindentation testing was carried out with a
diamond Berkovich tip at a maximum load of 1 mN with
loading and unloading rate of 200 lN/s and a hold time of
1 s. For coatings, profile indentation was carried out with
Ubi 3 nanoindentation system (Hysitron Incorporated,
Minneapolis, MN, USA). A total of 20-40 indents per
coating at 20-40 lm indent spacing were performed across
the center of the coating thickness cross-section. For
feedstock powders, splats and multi-pass splats, nano-
indentation mapping was conducted with a Triboindenter
(Hysitron Inc., Minneapolis, MN, USA) by indenting a
matrix of over 100 indents with lateral (x-y) indent spac-
ing of 2.5-3.0 lm. Data analysis and graphing software
Volume 20(3) March 2011—487
 (Origin 8.1, OriginLab Corp., Northampton, MA) was
Peer Reviewed
used to reconstruct the hardness and reduced modulus
distribution in terms of a color scale (i.e., maps of hardness
and modulus).
The hardness and elastic modulus of the materials were
calculated from the unloading curves using the Oliver and
Pharr method (Ref 26). The method is based on the
analysis of the unloading curve (load versus displacement)
obtained during sample indentation with a calibrated
indenter. For studies presented here, both the machine
compliance and the tip area function were determined
from indentation on a fused quartz standard.
The material hardness was calculated from Eq. 1,
where P is the maximum load applied by the tip and Ac is
the contact area of the tip.
P
H¼ ðEq 1Þ
Ac
The slope, dP/dh, of the unloading segment defines the
material stiffness, S, as well as the reduced elastic modu-
lus, Er (Ref 26, 27).
pffiffiffiffiffiffi
dP 2Er Ac
S¼ ¼ pffiffiffi ðEq 2Þ
dh p case, a region where the jetting region remains attached is
The contact area, Ac, in Eqs. 1 and 2 was determined from
the calibrated area function of the tip and a contact depth
determined from the Sneddon relationship used as part of
the Oliver and Pharr analysis (Ref 26).
The reduced elastic modulus consists of an inverse of
the sum of Poisson ratios of the material, tm, and indenter,
ti, divided by their respective elastic modulus, Em and Ei,
   1
1  t2i 1  t2m
Er ¼ þ ðEq 3Þ
Ei Em
With the indenter being diamond (Ei = 1120 GPa and
ti = 0.07) and the tested material being Ti (tm = 0.33), one
finds that the YoungÕs modulus of the tested material, Em
is nearly identical to the reduced modulus, Er. For
example, if Er = 100 GPa, Em = 102 GPa. The difference
between these two values is much less than the standard
deviation on the nanoindentation measurements for Er.
All the values for modulus reported in this article are
reduced modulus. Trends or differences in reduced mod-
ulus directly reflect trends or differences in the test
materialÕs YoungÕs modulus.
3. Results and Discussion
3.1 Microstructure of Cold Sprayed Splats
The appearance of cold spray splats was found to vary
as a function of deposition velocity, revealing features
distinctive to the deposition mechanisms. Representative
SEM images of splats deposited at velocities between 579
and 825 m/s (see Fig. 3) all showed material jetting, indi-
cating the presence of adiabatic shear instability (Ref 4).
However, the jetting phenomenon increased with increase
in the splat deposition velocity.
488—Volume 20(3) March 2011
The increase in material deformation and material
jetting with deposition velocity was further revealed by
examination of splat cross-sections (see Fig. 4a-c). As
deposition velocity increases, the flattening ratio increases
(see Fig. 4d). For a nondeformed powder particle the
flattening ratio is 1. The flattening ratio of the splats
deposited over a range of 570 to 825 m/s increases from
1.8 to 4. The increase from 1 to 1.8 appears to be largely
due to material deformation upon impact (minimal jetting
at lower velocities), while the increase from 1.8 to 4 seems
to be primarily associated with an increase in material
jetting. Aside from material jetting other interesting fea-
tures were observed that were due to the interaction of Ti
particles. At a deposition velocity of 636 m/s (see Fig. 5), a
crater formation can be seen at the impact site of the splat.
The impact of the splats resulted in a plastic deformation
in shape of the impacted body, typical of deformation
contributing to a conformal splat adhesion mechanism.
While there is also material jetting evident for this splat,
the splat has pulled away and partially de-bonded. At
higher deposition velocities, this de-bonding was not as
pronounced. In Fig. 6, a splat deposited at 770 m/s also
appears to have partially pulled away. However, in this
observed. While impact angle and location of impact has
an effect on these observations, with increasing deposition
velocity, the number of events resembling Fig. 6 generally
increased and the number of events resembling Fig. 5
decreased. Thus, the increase in jetting with deposition
velocity was a significant mechanism for bonding of par-
ticles in multi-pass splats.
At higher deposition velocity conditions, the jetting
phenomenon becomes more pronounced and evidence of
stronger splat adhesion is observed. Figure 6 shows a Ti
splat deposited on a previously deposited Ti splat at the
deposition velocity of 770 m/s. The close up of the splat
shows a continuous bond between two splats in jetting
region which could point to a formation of the metallur-
gical bond in that region (Ref 20).
3.2 Microstructure of Cold Sprayed Ti Coatings
Figure 7 shows cross-sectional images of three Ti
coatings that were produced at deposition velocities of
642, 724, and 825 m/s. The coating thickness ranged from
0.9 to 1.4 mm. The coating porosity decreases with
increase in the deposition velocity and dense coatings with
porosity below 2% are produced at deposition velocity
above 724 m/s. Figure 8(a) shows the effect of the depo-
sition velocity on the Ti coating porosity. The deposition
efficiency of the coatings increases with increase in the
deposition velocity and reaches 100% above the deposi-
tion velocity of 725 m/s. The transition coincides with the
increase in the flattening ratio observed in Fig. 4.
The coatings discussed in this article consist of single
line coatings where the particle impact angle is different
from standard multi-line coatings. Some differences in
coating porosity measurements and deposition efficiency
are expected between coatings produced here and those
produced at similar conditions, but as multi-line coatings
Journal of Thermal Spray Technology

Fig. 3 Micrographs (SEM) of CP Ti splats deposited at: (a) 300 C, 2 MPa (580 m/s), (b) 300 C, 4 MPa (642 m/s), (c) 500 C, 2 MPa
(636 m/s), (d) 500 C, 4 MPa (724 m/s), (e) 750 C, 3 MPa (770 m/s), and (f) 800 C, 4 MPa (825 m/s) gas preheat temperature, gas
pressure and particle velocity
(Ref 22). Single line coatings were chosen for this study
because the particle jet center was easily connected to the
center of the coating cross-section, avoiding any concerns
over variation of velocity within the jet affecting our
measurements (Ref 2).
3.3 Nanoindentation Mapping of Cold Sprayed
Splats and Multi-Pass Splats
Mechanical property mapping was carried out on the
cold spray Ti splats and feedstock Ti powder. Figure 9
depicts the distribution of hardness and modulus of feed-
stock Ti powder and cold spray splats deposited at 642,
724, and 825 m/s. The color notation is interpolated
between the measurements from indented data points.
The color scale has a gradient of 0.2 GPa for hardness and
10 GPa for modulus. The upper and lower limits of the
Journal of Thermal Spray Technology
Peer Reviewed
hardness and modulus color scale are annotated by gray
and black, while white indicates missing data points. In
Fig. 9, black regions represent the measurements made in
the epoxy which has low hardness, below 2.2 GPa, and
modulus, below 80 GPa. The regions in gray can be found
in the mild steel substrate that has a higher elastic mod-
ulus when compared to titanium. For comparison, the
optical images of the indented cold spray splats are shown
at the left-hand side of Fig. 9.
Nanoindentation maps reveal that the hardness distri-
bution inside of the cold spray splats or feedstock powder
is not homogeneous (see Fig. 9). In the feedstock powder,
high hardness of 3.6 to 3.8 GPa is measured at the
powder outer surface layer. A martensitic microstructure
was often observed in this region, possibly a result of
surface tension during the powder production through
a plasma atomization process. However, in splats,
Volume 20(3) March 2011—489
Peer Reviewed

Fig. 4 Micrographs (LOM) of Ti splat cross-section deposited at (a) 300 C, 4 MPa (642 m/s), (b) 500 C, 4 MPa (724 m/s), (c) 800 C,
4 MPa (825 m/s) gas preheat temperature, gas pressure, and particle velocity, and (d) the flattening ratio with standard deviation of Ti
splats with respect to the deposition velocity

Fig. 5 Micrographs (SEM) of splat deposited at 500 C, 2 MPa (636 m/s) and demonstrating a conformal particle adhesion

high hardness regions are mainly observed at the splat
impact site with the substrate material. The hardness in
these regions reaches 4.2 GPa and is color coded in red.
Relatively lower hardness of 3.0-3.4 GPa is measured in
the upper splat portion.
In Table 1, the average hardness and modulus of the
splats have been calculated. The average splat hardness
does not increase significantly with increase in the depo-
sition velocity. The splats hardness is, however, higher
than that of the feedstock powder. The modulus of the
single splats is similar to the modulus of the feedstock
powder. The residual stresses induced by impact are
expected to cause an increase in hardness but have little
effect on the elastic properties.
As opposed to the single pass splats, multi-pass splats
showed a higher modulus. In Fig. 10, modulus of 120 GPa
and higher is measured in the splats deposited above
500 C gas preheat temperature. As the gas preheat tem-
perature increases to 800 C, the modulus of the splats
increases up to 130 GPa. These values of the modulus are
consistent with that of Ti, and could simply be an effect of
differences in crystallographic texture.
In multi-pass splats, higher hardness is found both at
the impact site between splats and near the steel substrate.

490—Volume 20(3) March 2011 Journal of Thermal Spray Technology

 Peer Reviewed
Fig. 6 Micrographs (SEM) showing ductile fracture in the jetting region of CP Ti splat deposited at 750 C, 3 MPa (770 m/s)

Fig. 7 Micrographs (SEM) of cold spray Ti coatings deposited at: (a) 300 C, 4 MPa (642 m/s), (b) 500 C, 4 MPa (724 m/s), and
(c) 800 C, 4 MPa (825 m/s)

In Fig. 10, multi-pass titanium splats deposited at 642, 724,
and 825 m/s are shown along with hardness and modulus
maps. Regions of 3.7-4.2 GPa hardness were measured in
the impact region between two splats as well as in the
impact region of the splat on the steel substrate. However,
in the jetting region, a lower hardness of 3.2-3.4 GPa was
measured for the 825 m/s splat.
3.4 Nanoindentation of Cold Sprayed Ti Coatings
Profile nanoindentation across cold spray Ti coatings
revealed that the mechanical properties of these cold
spray coatings did not change with the coating thickness.
A typical profile of hardness and modulus measured
across coating thickness cross-section is shown in
Fig. 11(a) for a coating deposited at 500 C gas preheat
temperature, 4 MPa gas pressure, and 724 m/s particle
deposition velocity. Similar trends were observed for all
coatings produced at all deposition velocity conditions.
This result is contrary to previously published results for
microindentation on cold sprayed titanium (Ref 11, 22)
where the hardness was seen to increase near the sub-
strate/coating interface due to a tamping effect. This
difference is likely due to the effect porosity and particle

Journal of Thermal Spray Technology Volume 20(3) March 2011—491

Peer Reviewed

Fig. 8 Graphs of (a) porosity with a standard deviation and (b) deposition efficiency of cold spray Ti coatings vs. deposition velocity

Fig. 9 Hardness and modulus maps of Ti powder for Ti splats deposited at: 300 C, 3 MPa (642 m/s); at 500 C, 4 MPa (724 m/s), and at
800 C, 4 MPa (825 m/s). Micrographs (LOM) of the splats and powder are provided on the left-hand side. The hardness and modulus color
scale bars are provided at the right-hand side with black demonstrating lower end values and gray showing higher end values

de-bonding has on the hardness test. Nanoindentation is
affected the least and represents mechanical properties of
the cold sprayed material, while microindentation pro-
vides a measure of the overall mechanical response of the
coating, including defects.
The average hardness and modulus from profile
indentation is plotted against the deposition velocity in
Fig. 11(b), with the mechanical properties of the feedstock
Ti powder plotted at 0 m/s. Higher hardness and modulus
are measured in the cold spray coatings in comparison to
the titanium feedstock powder. The hardness and modulus
of the coatings deposited above 625 m/s was 3.7 ± 0.2
GPa and 125 ± 10 GPa, respectively. The hardness and
modulus of the initial feedstock powder was 3.2 ± 0.2 GPa
and 110 ± 5 GPa. No significant change in the coating
hardness with increase in the coating deposition velocity

492—Volume 20(3) March 2011 Journal of Thermal Spray Technology

Table 1 Mechanical properties of Ti powder and cold spray Ti coatings and Ti splats

Peer Reviewed
Deposition Deposition Coating Coating Coating Splat Splat Splat
velocity, efficiency, porosity, hardness, modulus, flattening hardness, modulus,
Specimen/conditions m/s % % GPa GPa ratio GPa GPa

Ti powder/as-polished N/a N/a <1 3.1 ± 0.3 107 ± 9 N/a 3.1 ± 0.3 107 ± 9
Ti/300 C, 2 MPa 580 19 18.0 ± 10.1 3.2 ± 0.3 99 ± 11 1.8 ± 0.1 - -
Ti/300 C, 3 MPa 625 30 2.5 ± 0.3 3.6 ± 0.2 112 ± 12 2.0 ± 0.2 3.4 ± 0.4 103 ± 9
Ti/300 C, 4 MPa 642 49 5.9 ± 1.5 3.7 ± 0.2 116 ± 13 2.2 ± 0.3 - -
Ti/500 C, 2 MPa 636 76 7.4 ± 1.6 3.6 ± 0.4 117 ± 10 2.4 ± 0.5 - -
Ti/500 C, 3 MPa 694 83 1.4 ± 0.6 3.7 ± 0.3 120 ± 11 2.5 ± 0.5 - -
Ti/500 C, 4 MPa 724 91 2.1 ± 0.6 3.5 ± 0.3 109 ± 8 2.3 ± 0.4 3.6 ± 0.4 119 ± 6
Ti/750 C, 3 MPa 770 100 0.7 ± 0.1 3.7 ± 0.4 120 ± 7 3.0 ± 0.6 - -
Ti/800 C, 4 MPa 825 100 1.3 ± 0.3 3.5 ± 0.3 121 ± 6 3.9 ± 1.0 3.7 ± 0.4 110 ± 7

Fig. 10 Hardness and reduced modulus maps of the multi-pass splats deposited at 300 C, 3 MPa (642 m/s); at 500 C, 4 MPa (724 m/s),
and at 800 C, 4 MPa (825 m/s). Micrographs (LOM) of the splats and powder are provided on the left-hand side. The hardness and
modulus color scale bars are provided at the right-hand side with black demonstrating lower end values and gray showing higher end
values

past 650 m/s range was observed. Average hardness 3.5 Deposition Mechanisms, Mechanical
and modulus for all the coatings are summarized in Properties, and Comparisons to Literature
Table 1.
Splat bonding mechanisms, extent of deformation,
It should be noted that mechanical property mapping
degree of jetting, and mechanical properties were all
was also conducted on coatings. Similar small-scale vari-
observed to vary to greater or lesser degrees as a function
ation in mechanical properties across particle boundaries
of deposition velocity. The relationships between all these
was also observed. Thus, while the average properties of
phenomena are complex, but the results of this study
the cold sprayed coatings do not change over the length
along with those in the literature leads to the following
scale of the coating thickness, small variations similar to
discussion of the observable trends with deposition
those observed in the splats do exist in the coatings at a
velocity.
similar length scale to the particles.

Journal of Thermal Spray Technology Volume 20(3) March 2011—493

Peer Reviewed

Fig. 11 (a) Nanoindentation profile hardness and modulus of Ti coatings deposited at 500 C, 4 MPa (724 m/s) away from the coating
and the substrate interface plotted at 0 mm, (b) Average profile nanoindentation hardness and modulus with standard deviation of
coatings plotted against their respective deposition velocities. For reference, the average properties for the powder are plotted at 0 m/s

When titanium was sprayed at velocities below 725 m/s,

the deposition efficiency was below 100% and the flat-
tening ratio of the splats was roughly constant between 1.8
and 2.5. While some material jetting was observed for
these splats, it was somewhat minimal compared to those
deposited at higher velocities (c.f. Fig. 3, 9). Additionally,
evidence of de-bonding of splats (c.f. Fig. 5) indicated that
the jetting was in some instances not providing a strong
metallurgical bond. Above 725 m/s, the deposition effi-
ciency was 100% and the flattening ratio rose to a value of
3 at 770 m/s and 3.9 at 825 m/s. In this regime, significant
jetting occurred and evidence metallurgical bonding was
also found (c.f. Fig. 6). These observations, especially the
increase in material jetting with particle velocity, are
consistent with the current understanding of the mecha-
nism of adiabatic shear instability as a deposition mecha-
nism for cold spray (Ref 4, 6, 7). The transition to 100%
deposition efficiency at roughly 725 m/s matches very well
the predicted critical velocity of ~700 m/s for 25 lm
diameter titanium particles, Schmidt et al. (Ref 4). In
terms of mechanical properties, the cold spray splats and
coatings are always on average harder than the feedstock
Ti powder. However, no significant increase in the coating
or splat hardness was measured with the increase in par-
ticle velocity. To explore this phenomenon further, we
plot the hardness of cold spray splats near the interface
and far from the interface as a function of deposition
velocity (see Fig. 12). Near interface hardness is generally
significantly higher than the feedstock powder, indicating
work hardening during the impact. This hardness remains
relatively constant with particle velocity, seeming to
indicate that the extent of work hardening has reached a
maximum in this area even at the lowest velocity. How-
ever, as the velocity increases, the hardness far from the
interface does increase gradually, perhaps due to some
deformation and increased dislocation density in the
upper splat regions. Lastly, the hardness of the jetting
region is also plotted in Fig. 12 for the splat with the
highest deposition velocity of 825 m/s (c.f. Fig. 10). In the
jetting region, the hardness is similar to the feedstock
powder. Thus, while the average mechanical properties
remain constant with deposition velocity, the complex
phenomenon of particle impact, deformation and jetting is
Fig. 12 Hardness distribution with respect to the splat deposi-
tion velocity in Ti splat regions: near interface, in the jetting
region and in the upper splat region
also evident in the variation of hardness in the different
regions of the splats. This discussion and the splat hardness
maps correlate well with the dislocation density distribu-
tion in the TEM image of Ti splats taken by Kim et al.
(Ref 24) and also the simulated stress distribution patterns
proposed by Assadi et al. (Ref 7) Thus, there appears to be
a consistent picture of high dislocation density and hard-
ness near the interface, minimal activity in the upper splat
region, and the possibility of thermal softening in the
jetting regions.
According to Kapoor et al., high strain rate deforma-
tion of Ti, such as that occurring in the jetting region, can
be very efficient at converting deformation work into heat
(Ref 25). The temperature rise in the jetting region was
estimated to reach near melting point (<90% of Tm) for Ti
splats deposited on a steel substrate with nitrogen gas at
950 m/s deposition velocity (Ref 20). Higher temperatures
have been estimated for Ti splats deposited on a Ti sub-
strate (Ref 28). The extent to which this heat affects the
resulting microstructure and distribution of strain is diffi-
cult to predict. Others have shown that the temperature
rise during impact of cold sprayed material is sufficient to
induce a dynamic recrystallization in the cold spray
deposited Ti splats (Ref 24). In addition, there has been

494—Volume 20(3) March 2011 Journal of Thermal Spray Technology

evidence that the recrystallization can result in preferen-
tial grain orientation of cold spray Ti coatings (Ref 14).
For all splats and coatings studied here, the modulus
measured by indentation is within the normal range for Ti,
which can vary from 100 to 145 GPa (c.f. Eq. 3). As it is
not a pure uniaxial test, indentation is not very sensitive to
changes variations due to crystallographic orientation
(Ref 29, 30). Thus, the small changes observed in our data
may or may not reflect some degree of preferred orien-
tation in the sprayed material.
4. Conclusions
The mechanical properties of cold sprayed Ti splats and
coatings were studied by nanoindentation. When com-
pared to the hardness of the feedstock powder, the dis-
tribution of hardness within individual Ti cold spray splats
demonstrated three regions: (1) an impact site with an
increase in hardness, (2) an upper splat region with similar
hardness to feedstock material, and (3) jetting regions
where the hardness is similar to the feedstock material.
At lower deposition velocities (<650 m/s), the first two
regions were more evident than the third, indicating pri-
marily conformal adhesion. At higher deposition veloci-
ties (>650 m/s), the third region, the jetting region,
became more pronounced, indicating the formation adia-
batic shear instability. The mechanical properties mea-
sured in these three regions are consistent with results in
the literature that show increased dislocation density near
the interface and dynamic recrystallization in the jetting
region. Thus, at the highest deposition velocities, which
are preferred due to low coating porosity and high depo-
sition efficiency, there was sufficient heat in the jetting
regions to promote dynamic recrystallization in Ti and a
reduction in hardness. However, there is retention of work
hardening at the impact site away from the jetting region,
where there is high strain, but insufficient strain rate to
realize a significant temperature rise. Despite the signifi-
cant differences in hardness in the different regions, the
average hardness of cold sprayed Ti splats remained
similar for all deposition velocities and was greater than
the feedstock material. The same was observed for the
cold spray coatings produced, where there were local
variations in mechanical properties but the coating
mechanical properties remained constant across the
coating thickness cross-section.
Acknowledgments
Financial support from the Canadian Foundation for
Innovation is gratefully acknowledged; the cold spray
equipment was provided by CFI project No. 8246 while
the nanoindentation equipment was provided by CFI,
LeaderÕs Opportunity Fund, project No. 13029. Opera-
tional funding for this project was provided by the Natural
Sciences and Engineering Research Council (NSERC)
Strategic Grants Program. The authors acknowledge the
Journal of Thermal Spray Technology
technical assistance of Ahmad Rezaeian, Bernard Harvey
Peer Reviewed
and Frederic Belval.
References
1. E. Irissou, J.G. Legoux, A. Ryabinin, B. Jodoin, and C. Moreau,
Review on Cold Spray Process and Technology: Part I, Intellec-
tual Property, J. Therm. Spray Technol., 2008, 17(4), p 495-516
2. A. Papyrin, V. Kosarev, K. V. Klinkov, A. Alkhimov, and
V. M. Fomin, Cold Spray Technology, Elsevier, Oxford, 2006,
p 74, 153-169, 259
3. V.K. Champagne, The Cold Spray Materials Deposition Process:
Fundamentals and Applications, Woodhead Publishing Limited,
Cambridge, 2007, p 362
4. T. Schmidt, F. Gärtner, H. Assadi, and H. Kreye, Development
of a Generalized Parameter Window for Cold Spray Deposition,
Acta Mater., 2006, 54(3), p 729-742
5. C.-J. Li, W.-Y. Li, and H. Liao, Examination of the Critical
Velocity for Deposition of Particles in Cold Spraying, J. Therm.
Spray Technol., 2006, 15(2), p 212-222
6. M. Grujicic, C.L. Zhao, W.S. DeRosset, and D. Helfritch, Adi-
abatic Shear Instability Based Mechanism for Particles/Substrate
Bonding in the Cold-Gas Dynamic-Spray Process, Mater. Des.,
2004, 25(8), p 681-688
7. H. Assadi, F. Gärtner, T. Stoltenhoff, and H. Kreye, Bonding
Mechanism in Cold Gas Spraying, Acta Mater., 2003, 51(15),
p 4379-4394
8. T. Schmidt, F. Gaertner, and H. Kreye, New Developments in
Cold Spray Based on Higher Gas and Particle Temperatures,
J. Therm. Spray Technol., 2006, 15(4), p 488-494
9. C. Borchers, F. Gartner, T. Stoltenhoff, and H. Kreye, Micro-
structural Bonding Features of Cold Sprayed Face Centered
Cubic Metals, J. Appl. Phys., 2004, 96(8), p 4288-4292
10. T. Kairet, M. Degrez, F. Campana, and J.P. Janssen, Influence of
the Powder Size Distribution on the Microstructure of Cold-
Sprayed Copper Coatings Studied by X-ray Diffraction, J. Therm.
Spray Technol., 2007, 16(5), p 610-618
11. A. Rezaeian, E. Irissou, R.R. Chromik, S. Yue, Characterization
of Cold-Sprayed Ni, Ti, Cu Coating Properties for their
Optimization, Thermal Spray: Crossing Borders, Jun 2-4, 2008
(Maastricht, The Netherlands), ASM International, p 854-859
12. T.H. Van Steenkiste, J.R. Smith, R.E. Teets, J.J. Moleski,
D.W. Gorkiewicz, R.P. Tison, D.R. Marantz, K.A. Kowalsky,
W.L. Riggs, P.H. Zajchowski, B. Pilsner, R.C. McCune, and
K.J. Barnett, Kinetic Spray Coatings, Surf. Coat. Technol., 1999,
111(1), p 62-71
13. T. Stoltenhoff, C. Borchers, F. Gärtner, and H. Kreye, Micro-
structures and Key Properties of Cold-Sprayed and Thermally
Sprayed Copper Coatings, Surf. Coat. Technol., 2006, 200(16-17),
p 4947-4960
14. S. Zahiri, D. Fraser, and M. Jahedi, Recrystallization of Cold
Spray-Fabricated CP Titanium Structures, J. Therm. Spray
Technol., 2009, 18(1), p 16-22
15. C.-J. Li and W.-Y. Li, Deposition Characteristics of Titanium
Coating in Cold Spraying, Surf. Coat. Technol., 2003, 167(2-3),
p 278-283
16. W.Y. Li, C. Zhang, X. Guo, J. Xu, C.J. Li, H. Liao, C. Coddet,
and K.A. Khor, Ti and Ti-6Al-4V Coatings by Cold Spraying and
Microstructure Modification by Heat Treatment, Adv. Eng.
Mater., 2007, 9(5), p 418-423
17. R.S. Lima, A. Kucuk, C.C. Berndt, J. Karthikeyan, C.M. Kay, and
J. Lindemann, Deposition Efficiency, Mechanical Properties and
Coating Roughness in Cold-Sprayed Titanium, J. Mater. Sci.
Lett., 2002, 21(21), p 1687-1689
18. T. Marrocco, D. McCartney, P. Shipway, and A. Sturgeon, Pro-
duction of Titanium Deposits by Cold-Gas Dynamic Spray:
Numerical Modeling and Experimental Characterization,
J. Therm. Spray Technol., 2006, 15(2), p 263-272
19. J. Vlcek, H. Huber, H. F. Voggenreiter, and E. Lugscheider,
Melting Upon Particle Impact in the Cold Spray Process, Inter-
national Congress on Advanced Materials, Their Process and
Volume 20(3) March 2011—495
 Applications, September 30-October 2, 2002 (Munich, Germany),
Peer Reviewed
Deutsche Gesellschaff Fur Materialkunde (DSM)
20. G. Bae, S. Kumar, S. Yoon, K. Kang, H. Na, H.J. Kim, and C. Lee,
Bonding Features and Associated Mechanisms in Kinetic Sprayed
Titanium Coatings, Acta Mater., 2009, 57(19), p 5654-5666
21. K. Kim, M. Watanabe, K. Mitsuishi, K. Iakoubovskii, and S. Kuroda,
Impact Bonding and Rebounding Between Kinetically Sprayed
Titanium Particle and Steel Substrate Revealed by High-Resolution
Electron Microscopy, J. Appl. Phys. D, 2009, 42(6), p 65304
22. W. Wong, A. Rezaeian, E. Irissou, J.G. Legoux, and S. Yue, Cold
Spray Characteristics of Commercially Pure Ti and Ti-6Al-4V,
Adv. Mater. Res., 2010, 89-91, p 639-644
23. K.-H. Kim, M. Watanabe, J. Kawakita, and S. Kuroda, Grain
Refinement in a Single Titanium Powder Particle Impacted at
High Velocity, Scripta Mater., 2008, 59(7), p 768-771
24. K.-H. Kim, M. Watanabe, J. Kawakita, and S. Kuroda, Effects of
Temperature of In-flight Particles on Bonding and Microstruc-
ture in Warm-Sprayed Titanium Deposits, J. Therm. Spray
Technol., 2009, 18(3), p 392-400
496—Volume 20(3) March 2011
25. R. Kapoor and S. Nemat-Nasser, Determination of Temperature
Rise During High Strain Rate Deformation, Mech. Mater., 1998,
27(1), p 1-12
26. A. C. Fisher-Cripps, Nanoindentation, 2nd ed., Springer,
New York, 2002, p 1-20
27. W.C. Oliver and G.M. Pharr, Measurement of Hardness and
Elastic Modulus by Instrumented Indentation: Advances in
Understanding and Refinements to Methodology, Mater. Res.
Soc., 2004, 19(1), p 3-20
28. G. Bae, Y. Xiong, S. Kumar, K. Kang, and C. Lee, General
Aspects of Interface Bonding in Kinetic Sprayed Coatings, Acta
Mater., 2008, 56(17), p 4858-4868
29. J.J. Vlassak and W.D. Nix, Measuring the Elastic Properties of
Anisotropic Materials by Means of Indentation Experiments,
J. Mech. Phys. Solids, 1994, 42(8), p 1223-1245
30. J.J. Vlassak, M. Ciavarella, J.R. Barber, and X. Wang, The
Indentation Modulus of Elastically Anisotropic Materials for
Indenters of Arbitrary Shape, J. Mech. Phys. Solids, 2003, 51(9),
p 1701-1721
Journal of Thermal Spray Technology